nakamochi
/
3d
Archived
2
0
Fork 0
Commit Graph

29 Commits (bbc38e6c3382c19d5ca9748838a5eb9b375fd946)
 

Author SHA1 Message Date
alex bbc38e6c33 nerd/box: add a box mfr tolerance parameter
earlier, some dimensions were sized as dy + a hardcoded 0.5mm as a way
to adjust for FDM inaccuracy.

with this commit, the adjustment is now moved to a new parameter
box_mfr_tolerance so that it's easier to prepare for different tech
process: FDM, SLA, jet, etc.

the parameter is now set to 0mm because we're moving towards more
accurate processes, SLA and jet. but it is always possible to set this
back to 0.5mm or another value and generate an STL mesh file.
just watch out for freecad's topo naming problem.
10 months ago
alex 5357a7b3c9 nerd/box: add wallstrips reinforcement
a couple of strips on the long sides of the box walls to keep them
straight.
10 months ago
alex cc01b93f88 nerd: actually rename diyplus.FCStd main file to nerd
missed it in the previous commit.
10 months ago
alex d29cb71e90 delete fff version and rename diyplus into "nerd"
this is to align with the naming scheme we came up on the website at
https://nakamochi.io/#editions

essentially, we merged fff, aka DIY, and DIY+ which lived in diyplus
directory, into one called NERD. this is because the simplest DIY didn't
provide acceptable performance characteristics and they are both
suitable for a home-made DIY project anyway. hence, the "nerd" edition
aka "have fun playing around with it on your own".

it also makes things simpler. the other version we'll be making is KAFI.
there are no source files for it yet.
10 months ago
alex db62a7be82 diyplus/ssdsupport: reduce SSD mount screw holes from 3mm to 2mm
the screws are actually about 2.5mm in diameter, reused from SSD
teardown. they are wide threaded, for plastic.

3mm hole diameter was to accomodate for FDM 3D printers inaccuracies but
it has shown to be too large, especially when produced with SLA or SLS
method.
10 months ago
alex ffff9ea60d remove generated files - leave only sources
any time FCStd sources are changed, the generated STL, STEP, DXF and
other derivatives are no longer current. it makes more sense to post and
tag the latter in releases.
10 months ago
alex bf7d7c5e29 readme: make markdown table syntax valid for both gitea and github 2 years ago
alex fe29892c02 fff: update readme with a more detailed BOM and a build photo 2 years ago
alex d9d8f89b7a diyplus: a new and improved version compared to FFF
this version features a custom heatsink combined with the bottom plate, suitable
for aluminium CNC milling. the construction provides a better heat dissipation
compared to the DIY (FFF) version.

see all details in the readme.
2 years ago
alex e92ee1f432 fff: assembly cosmetic updates; no changes to the enclosure
just shuffling parts around
2 years ago
alex 9dc2a5e91c fff: small final tweaks; stl files
successfully printed a couple cases using 1.75 PLA, 0.4mm head,
and assembled into a working prototype. all seem to fit together ok.
2 years ago
alex 714fc4b4df fff: replace fused parts with a single body
the fuse function in freecad didn't seem to work well when trying to go
back and update individual body parts. this commit is essentially a
remake of the plate + wall parts fuse from 6309e76 as a single body.

the reason to go back and modify individual parts was to move SSD for
more estate, to fit a 90deg connector because a straight cable doesn't
fit.

while there, added missing screws and spacers to the assembly.
this version should be good to start 3d printing.
2 years ago
alex 6f459e39c1 assets: add a simple 90degree usb-c adapter
turned out the cable connector attached to SSD is typically too long to
fit. this 90deg adapter helps.

the SSD is going to be connected to the main board via this adapter
in the next commit.
2 years ago
alex 6309e766b3 fff: fuse plate + walls into a new body and chamfer everywhere
some first 3d print trials reveal the box might not be sufficiently strong
and easy to print. adding chamfers and fillets helps.

adding chamfers and fillets on plate/wall edges is much easier when the
thing is a single body.
2 years ago
alex 531d178044 fff: let air flow from sidecup bottom through LCD wall left face 2 years ago
alex f2c0c72e16 fff: add power cord and sd card access slots at the rear face 2 years ago
alex 27f6c9104f fff: add a side compartment part to close the gap
the part is called sidecup. this is where cables reside, and possibly
a custom PCB or a battery pack in the future.
2 years ago
alex df29aa2bfd fff: combine walls and the bottom plate to form a box
makes it easier to 3d-print in the end.

the plate and walls are still separate parts - easier to work on
separately. it is also possible to even 3d-print them separately, in
which case add a 0.5mm to clearance and plate mount Z in the params
sheet.
2 years ago
alex aceae05b6e fff: correct rpi board and top heatsink Z in provcase assembly
were off by ~3mm lower. this will actually make a good Z for top/bottom
heatsink spacer of about 5-6mm.
2 years ago
alex 81440fd637 fff: add walls around LCD to provcase
with a standard M2 30mm PCB standoff for LCD, the walls' Z seems
to magically settle on a nice round number of 48mm.
2 years ago
alex 7c73301bd7 fff: improve and split the assembly into a separate file
this makes it a lot easier to work on a part and look at it in the
assembled view by simply switching tabs. the separate file links to
the original provcase.FCStd. so, need both files to work on the assembly.

it is now possibly to simply select "exploded" or "assembled" option
from a "view" property of the top level assembly std part: no need to
manually modify the asmsheet. the property config is in the same
spreadsheet.

display standoff length now correctly offsets the LCD, the board and top
heatsink while leaving the rest in place. so, it is very easy to change
the overall height.

this change also applies 4.3inch LCD updates from 0c42295 with correct
rpi board mount screw holes, since the LCD master sketch in the
assembly is a copy.
2 years ago
alex 0c42295ff2 assets: correct waveshare 4.3inch LCD rpi board mount points
turns out either the original design is inaccurate or the freecad STEP
import didn't work well: the rpi mounting screw holes were off but some
sub millimeter length: 58.2x49.05 vs 58x49 actual.

see rpi drawings in
https://datasheets.raspberrypi.com/rpi4/raspberry-pi-4-mechanical-drawing.pdf
2 years ago
alex 654d843983 fff: adjust base plate to match the SSD PCB mount pins
the SSD PCB sits very nicely now. the heatsink model has a pad at the
bottom but it is not the case with the actual sink in use.
will remove it later.

the change also makes the base plate shorter, right about the size of
the display: will be splitting the box in two to allow printing within a max
circle area of 140mm diameter.
2 years ago
alex 8890e1f618 assets: add a master sketch to waveshare 4.3inch LCD
this also means the STEP file is replaced with a freecad native file
format.

the master sketch comes in very handy when designing mount and other
constructions around the LCD.
2 years ago
alex 9f082cd5f4 fff: add the sdssde30-1t00 pcb to the assembly
it shows in fact that the ssd pcb mounting pins in the provcase bottom
part are slightly misplaced. will adjust in future commits.
2 years ago
alex 1acdb2f30f assets: add a simple model of sdssde30-1t00 pcb
very simple, contains only the essential bits dimensions of which play
important role in the case design due to size restrictions.

there are two std parts: regular with sketches with sketches to
construct the body, and a simple refined version for the provcase
assembly.

the regular variant also has a master sketch to aid in defining
constraints in the provcase design.
2 years ago
alex ddc0711c74 fff: add bottom sink screws to the assembly
the standard hex with low head M3x10; comes with the display pkg.
2 years ago
alex 310795898a fff: align all parts in the assembly
all are parameterized via the asmsheet which now includes
a calculation for an exploded view.

while there, added the missing display standoffs.

what's still missing is:

- SSD board
- heatsink top/bottom spacers
- outer case walls and mount holes
2 years ago
alex 928bab8f0c initial import
certainly not ready for broad consumption yet.
just sharing very early designs with friends.
2 years ago